KManager V4.1.4

K-Series Programmable ECU installation questions / support issues
User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

KManager V4.1.4

Post by Hondata » Thu Apr 30, 2015 1:32 pm

V4.1.0 (30 Apr 2015)
  • Changed ignition table adjustment so that 'Set Values' applies the same value to all cam angle tables if 'Edit All Tables' is selected.
  • Added function selection to nitrous outputs
  • Moved Closed Loop Operation to Closed Loop tab.
  • PRA fixed constant speed setting.
  • [KPro4 / PRA & PRB] Added Closed Loop Target Lambda tables
V4.1.1 (13 May 2015)
  • USB drivers updated to v1.0.3 (Microsoft Hardware Certified)
  • [KPro4] Added flex fuel ethanol sensor smoothing and max rpm
V4.1.2 (30 June 2015) V4.1.3 (2 July 2015)
  • [KPro4] Fixed PRA target lambda staying at 14.7
V4.1.4 (27 July 2015)
  • [KPro4] Fixed bug where live tuning would not update when multiple cam angles of the same table were selected.
  • Added analog input conversion for the Delphi 12160855 / AEM 30-2012 fluid temperature sensor.
Download Here

The main change is the addition of target lambda tables and the ability to run closed loop under boost. See the help file for instructions on how to do this.

The USB driver has been certified by Microsoft so will install on Windows 7 and above without needing confirmation. It also is available in Windows Update.
Hondata

Arno
Posts: 41
Joined: Sun Sep 02, 2012 8:41 am
Location: The Netherlands

Re: KManager V4.1.0 (beta)

Post by Arno » Sat May 02, 2015 2:30 am

Hi,

There seems to be a small bug in the GUI code or .kal parsing in 4.1.0 compared to 4.0.8 as far as I can tell.

I use a PRA ECU and calibrations and with 4.1.0 some options in the Parameters section get greyed out and become un-selectable when starting Kmanager and opening an existing KAL. Especially options like like 'Use stock control logic' under Knock/TPS and Fuel pump 'Priming time' under Misc seem to be affected.

Sometimes closing and re-opening the calibration (but not stopping Kmanager) makes them selectable/settable again or selecting a default calibration and then closing that and re-opening the original one helps, but this is not always the case. This under with Win 7 starter edition on an Asus EeePC.

Downgrading to 4.0.8 returns everything back to normal and no greying out of options occurs.

This is purely the Kmanager GUI or internals as far as I can tell. No ECU interaction at all. Without an ECU connected it also occurs.

Bye, Arno.

User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: KManager V4.1.0 (beta)

Post by Hondata » Sat May 02, 2015 8:17 am

The stock knock control logic is coupled to VTC being enabled; there was a bug which I have fixed.

The fuel pump prime time works ok as far as I can tell. Be aware that some options are only available on some calibration types and KPro generations. If KManager cannot tell what KPro is being used the option will be enabled, but it may change once a KPro is plugged in.
Hondata

D-Rob
Posts: 61
Joined: Tue Apr 29, 2008 6:11 am

Re: KManager V4.1.0 (beta)

Post by D-Rob » Sun May 03, 2015 6:45 am

The closed loop targets under high load/boost sounds cool, especially with the ethanol content sensor stuff, but how fast does it correct?

Has this been tested on high power, high boost applications at all?

Thanks,
Derek

User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: KManager V4.1.0 (beta)

Post by Hondata » Sun May 03, 2015 8:36 am

The correction time depends on a whole bunch of factors, but it helps to have the tables close to begin with and rely on the closed loop to make the initial corrections for a track, then correct the overall fuel. I was seeing about 2 cycles per second after a gearshift, but did not test enabling the fuel overrun cut delay so that the lambda would not go lean during the shift. After about 1.5-2 seconds from a gearshift the AF was really close to target - no more than 0.1 off.
High hp - not yet (~430 hp), but if Javier tests this weekend we'll have data from 1000+ hp.
Hondata

D-Rob
Posts: 61
Joined: Tue Apr 29, 2008 6:11 am

Re: KManager V4.1.0 (beta)

Post by D-Rob » Mon May 04, 2015 5:53 am

Thank you for the information.

Will switching from the fixed WOT compensation values to the new Target AFR tables require retuning of the fuel tables for use with the new Closed Loop features?

User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: KManager V4.1.0 (beta)

Post by Hondata » Mon May 04, 2015 11:26 am

It depends if you had WOT compensation tables enabled previously. If you did, then since the engine is in closed loop you'll need to apply the WOT compensation to the fuel tables.
Hondata

D-Rob
Posts: 61
Joined: Tue Apr 29, 2008 6:11 am

Re: KManager V4.1.0 (beta)

Post by D-Rob » Tue May 05, 2015 5:18 am

Forgive me, you're saying if the tables were used in the configuration attached, that using the new closed loop option will require retuning of the fuel tables themselves?

The car is already tuned to AFR targets very close to the new "Target Lambda" native settings, so would enabling them require extensive changes to the fuel tables?

I apologize for the barrage of questions.

Thank you!
Attachments
WOTCompQuestion.png
WOTCompQuestion.png (37.27 KiB) Viewed 9069 times

User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: KManager V4.1.0 (beta)

Post by Hondata » Tue May 05, 2015 6:30 am

Not re-tuning, but adjustment, as the effect can be easily calculated.
If you uncheck fix WOT comp to 0.85 and check target lambda, then the fuel tables will be 1/0.85 = 17.5% lean. So increase all the fuel tables under full load by that amount. You can also blend in the WOT compensation so that there is not a big step in AF when the ECU goes into WOT mode. eg columns 10 and above add 17.5%, column 9 10%, column 8 5%
I'm fairly sure I'm going to add the option to use the target lambda tables as WOT compensation, so keep track of fuel changes in case you need to reverse them.

Edit: The above is not correct - when using target lambda tables, the fuel values are changed by the target lambda, both in open and closed loop. Thus if you switch from WOT compensation fixed at 0.85 to target lambda tables, then you do not need to make fuel changes if you make the target lambda tables 0.85 as well.
Hondata

Knox
Posts: 122
Joined: Mon Jun 06, 2011 6:01 pm

Re: KManager V4.1.0 (beta)

Post by Knox » Wed May 06, 2015 5:58 am

So, if I understand correctly. When "WOT compensation fixed at lambda 0.85" is checked and high load columns tuned in open loop correctly everything works perfect as before 4.1.0. But If uncheck "WOT compensation fixed at lambda 0.85" and check "enable target lambda tables" we need to add fuel under high load tables because ecu add fuels by 17.5% because of checked "WOT compensation fixed at lambda 0.85" even if we tuned the high load columns in open loop right?

Sorry I want to be clear because English is not my first language.

And one more question.
Hondata you wrote
[KPro4 / PRA & PRB] Added Closed Loop Target Lambda tables

Does jdm prc and prd ecu's can use this closed loop target lambda tables with kpro4 borad?

Thanks.



Hondata wrote:Not re-tuning, but adjustment, as the effect can be easily calculated.
If you uncheck fix WOT comp to 0.85 and check target lambda, then the fuel tables will be 1/0.85 = 17.5% lean. So increase all the fuel tables under full load by that amount. You can also blend in the WOT compensation so that there is not a big step in AF when the ECU goes into WOT mode. eg columns 10 and above add 17.5%, column 9 10%, column 8 5%
I'm fairly sure I'm going to add the option to use the target lambda tables as WOT compensation, so keep track of fuel changes in case you need to reverse them.

User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: KManager V4.1.0 (beta)

Post by Hondata » Wed May 06, 2015 8:56 am

Yes - your tuning summary is correct. I do plan to change this behavior though.
Edit : See my post below.

> Does jdm prc and prd ecu's can use this closed loop target lambda tables with kpro4 borad?

Target lambda was added for the PRA and PRB calibration types. The calibration type is independent from the ECU.
Hondata

Knox
Posts: 122
Joined: Mon Jun 06, 2011 6:01 pm

Re: KManager V4.1.0 (beta)

Post by Knox » Wed May 06, 2015 11:11 am

Ok, thank you for support.

Hondata wrote:Yes - your tuning summary is correct. I do plan to change this behavior though.

> Does jdm prc and prd ecu's can use this closed loop target lambda tables with kpro4 borad?

Target lambda was added for the PRA and PRB calibration types. The calibration type is independent from the ECU.

djarmyssg
Posts: 20
Joined: Sat May 22, 2010 3:50 am

Re: KManager V4.1.0 (beta)

Post by djarmyssg » Thu May 07, 2015 7:45 am

Another glitch... When looking for "New Calibration".. Calibrations seem to be pushed down beginning at the s2000. IE.. s2000 maps are in D17 dropdown.. d17 maps are in the OTHER dropdown.. and one set of d17 and LOTUS Cals are missing completely...

User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: KManager V4.1.0 (beta)

Post by Hondata » Thu May 07, 2015 9:37 am

Thanks for the report. I've fixed this, but can't see the Lotus calibration you mentioned.
Hondata

User avatar
Hondata
Site Admin
Posts: 8759
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Re: KManager V4.1.0 (beta)

Post by Hondata » Mon May 11, 2015 11:04 am

I've re-tested and some of the above information I posted is not correct.

When using target lambda tables, the fuel values are changed by the target lambda, both in open and closed loop. Thus if you switch from WOT compensation fixed at 0.85 to target lambda tables, then you do not need to make fuel changes if you make the target lambda tables 0.85 as well.

This way you can change the target lambda values and not need to re-tune.
Hondata

Post Reply